Healthcare analytics leverages Artificial Intelligence and Cloud Computing in disease prediction, Clinical Decisions Making, Personalized Treatment and Remote Monitoring. However, having all of the information in a single place increases the risk of medical information being breached, misused, accessed and new threats in the quantum era. This study aims at designing an AI-powered healthcare analytics system with privacy in post-quantum cloud computing era. Incorporates Federated learning, Differential privacy, Secure aggregation, Post quantum key encapsulation, Quantum resistant digital signature, Role based access control and Explainable Artificial Intelligence. Patients’ data is kept locally and a global model is trained collaboratively by healthcare institutions. Differential privacy preserves the contribution of the individual, secure aggregation presents the institutional update in an obscure manner and post quantum mechanisms secure authentication and communication. An evaluation based on a diabetes dataset demonstrates the framework’s ability to classify with 92.94% accuracy with a reduction in membership-inference and gradient-reconstruction risks while using simulation. The latency, memory consumption and communications overhead of security mechanisms are high, but they are kept under control.